High-Throughput Optimization of Paper-Based Cell-Free Biosensors

Abstract

Cell-free expression systems maintain core cellular processes without intact cells and offer attractive properties as point-of-need biosensors. The ability to lyophilize, store, and use on-demand make these sensors usable in the field, and the lack of membranes means that there are no analyte transport issues and that new sensors can be deployed by simply adding a different DNA molecule. The lack of membranes also means that sensor designs and reaction optimizations can be screened in high throughput. While shelf stability has been demonstrated in specific cases using additives, these approaches are not universal to the myriad cell-free expression methods and formats. Here, we present new high-throughput screening methods to optimize cell-free expression formulations when embedded into paper for use as sensors. Our method leverages acoustic liquid handling to dispense reactions onto 384-well paper ticket formats and machine vision to quantify reaction performance from a colorimetric reporter enzyme. The throughput enabled shifts the bottleneck from experimental execution to selecting the experiments to execute; we therefore implement design-of-experiments to optimize the information gained from each design-build-test-learn cycle. We used these approaches to first optimize the performance of a low-cost cell-free expression formulation that was initially non-functional when embedded in paper, then further optimize it for tolerance to exposure to heat. With only 2 rounds of experimentation lasting 4 days total for each goal, the result are an energy mixture with 8% of the materials cost of a commonly used version and a formulation of excipients that maintain 60% of activity after 6 hours of storage at 50 {degrees}C and. Finally, we showcase the use of the cost-optimized formulation in a 3D-printed paperfluidic device where it outperforms the standard formulation at much lower cost.